What is it about?
Model search or AutoML usually requires a large amount of training data. However, such training data is not available in many model deployment scenarios. Therefore, how to select a pre-trained model and deploy it without any training data becomes an important problem. This work provides a benchmark of multiple approaches to searching deep neural network models without any training data for deployment.
Featured Image
Photo by GuerrillaBuzz Crypto PR on Unsplash
Read the Original
This page is a summary of: Benchmark of DNN Model Search at Deployment Time, July 2022, ACM (Association for Computing Machinery),
DOI: 10.1145/3538712.3538725.
You can read the full text:
Contributors
The following have contributed to this page
